pao_model Module Reference

Module for equivariant PAO-ML based on PyTorch. More...

Functions/Subroutines
subroutine, public	pao_model_load (pao, qs_env, ikind, pao_model_file, model)
	Loads a PAO-ML model.
subroutine, public	pao_model_predict (pao, qs_env)
	Fills paomatrix_X based on machine learning predictions.
subroutine, public	pao_model_forces (pao, qs_env, matrix_g, forces)
	Calculate forces contributed by machine learning.

Detailed Description

Module for equivariant PAO-ML based on PyTorch.

Author

Ole Schuett

Function/Subroutine Documentation

pao_model_load()

subroutine, public pao_model::pao_model_load	(	type(pao_env_type), intent(in)	pao,
		type(qs_environment_type), intent(in)	qs_env,
		integer, intent(in)	ikind,
		character(len=default_path_length), intent(in)	pao_model_file,
		type(pao_model_type), intent(out)	model
	)

Loads a PAO-ML model.

Parameters

pao	...
qs_env	...
ikind	...
pao_model_file	...
model	...

Definition at line 64 of file pao_model.F.

Here is the call graph for this function:

Here is the caller graph for this function:

pao_model_predict()

subroutine, public pao_model::pao_model_predict	(	type(pao_env_type), pointer	pao,
		type(qs_environment_type), pointer	qs_env
	)

Fills paomatrix_X based on machine learning predictions.

Parameters

pao	...
qs_env	...

Definition at line 146 of file pao_model.F.

Here is the call graph for this function:

Here is the caller graph for this function:

pao_model_forces()

subroutine, public pao_model::pao_model_forces	(	type(pao_env_type), pointer	pao,
		type(qs_environment_type), pointer	qs_env,
		type(dbcsr_type)	matrix_g,
		real(dp), dimension(:, :), intent(inout)	forces
	)

Calculate forces contributed by machine learning.

Parameters

pao	...
qs_env	...
matrix_G	...
forces	...

Definition at line 180 of file pao_model.F.

Here is the call graph for this function:

Here is the caller graph for this function: